THE RADIATION OF THE SUN—ABBOT. 147 
’ 
for which he knows the length of path in air, he may compute from 
the observed increase of intensity, attending the decrease of air path, 
how much the intensity would be if the path in air could be reduced 
to nothing at all, or, in other words, if he could go outside the air 
altogether. It is not possible to do this by observation with the 
pyrheliometer alone, as explained above, because the rays of certain 
spectrum wave lengths are almost entirely removed in the upper 
atmosphere, and do not reach the observer at all, even if he be on a 
high mountain. Especially is this the case in the infra-red region of 
the spectrum, which is invisible to the eye, but which is of great 
importance as containing a large part of the sun’s energy. In this 
region there are great water-vapor bands, where the water vapor of 
the atmosphere almost completely absorbs the solar rays, leaving 
ereat gaps in the representation of the sun’s energy spectrum. Lang- 
ley introduced the procedure of estimating for all other parts of the 
spectrum the intensity which would be found outside the atmosphere, 
le Scale 30 ae a ie ee 
Fic. 3.—Bolographs of the Solar Spectrum. Air masses of observation: Upper curve, 3.0; middle curve, 
4.0; lower curve, 5.2. 
but in the great water-vapor and other terrestrial bands of absorption 
he merely made the assumption that these would be altogether absent 
if he could in fact be beyond the atmosphere altogether. 
After Langley became Secretary of the Smithsonian Institution 
he established the Astrophysical Observatory there, in order that he 
might carry out to greater perfection the measurements of radiation 
begun by him while still director of the Allegheny Observatory in 
Pennsylvania. Among the first improvements introduced in Wash- 
ington was the automatic recording of the results of the bolometer by 
photographic means. This was a great step, so that now we are able, 
in the lapse of less than 10 minutes, to observe the intensity of the 
rays of the sun of all wave lengths, from those far beyond the violet 
end of the visible spectrum to those far beyond its extreme red. 
Figure 3 shows the result of three such observations made on Mount 
Wilson, in California, at the station of the Smithsonian Institttion 
there. These three curves represent the distribution of solar radia- 
tion in its spectrum, including the ultra-violet, visible, and infra-red 
rays. ‘The great water-vapor bands above mentioned are shown 
